You tried to integrate 2 systems (front end and back end) with an HTTP interface to one large system. These subsystems don't store any state inside. All of the state information is stored in a DynamoDB table. You have launched each of the subsystems with separate AMIs. After testing, these servers stopped running and are issuing malformed requests that do not meet the HTTP specifications of the client. Your developers fix the issue and deploy the fix to the subsystems as soon as possible without service disruption. What are the 3 most effective options from the below to deploy these fixes and ensure that healthy instances are redeployed?
A. Use VPC.
B. Use AWS Opsworks autohealing for both the front end and back end instance pair.
C. Use Elastic Load balancing in front of the front-end system and Auto scaling to keep the specified number of instances.
D. Use Elastic Load balancing in front of the back-end system and Auto scaling to keep the specified number of instances.
E. Use Amazon Cloudfront with access the front end server with origin fetch.
F. Use Amazon SQS between the front end and back end subsystems.
